---
dataset_info:
  features:
  - name: user
    dtype: string
  - name: assistant
    dtype: string
  - name: topic
    dtype: string
  splits:
  - name: train
    num_bytes: 30533621.022576667
    num_examples: 85428
  - name: validation
    num_bytes: 3392981.977423331
    num_examples: 9493
  download_size: 16047922
  dataset_size: 33926603.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---